conversion of regional data between nuts...
TRANSCRIPT
López-Cobo M.
Adapting the RHOMOLO
database to different uses
Conversion of regional data between NUTS classifications
2016
EUR 28300 EN
This publication is a Technical report by the Joint Research Centre (JRC), the European Commission’s science and
knowledge service. It aims to provide evidence-based scientific support to the European policy-making process.
The scientific output expressed does not imply a policy position of the European Commission. Neither the
European Commission nor any person acting on behalf of the Commission is responsible for the use which might
be made of this publication.
Contact information
Name: Montserrat López-Cobo
Address: Edificio Expo. c/ Inca Garcilaso, 3. E-41092 Seville (Spain)
E-mail: [email protected]
Tel.: 34 9544 80570
JRC Science Hub
https://ec.europa.eu/jrc
JRC104030
EUR 28300 EN
PDF ISBN 978-92-79-64453-5 ISSN 1831-9424 doi:10.2791/039045
Luxembourg: Publications Office of the European Union, 2016
© European Union, 2016
Reproduction is authorised provided the source is acknowledged.
How to cite: López-Cobo M. (2016); Conversion of regional data between NUTS classifications. Adapting the
RHOMOLO database to different uses; EUR 28300 EN; doi:10.2791/039045
All images © European Union 2016
2
Table of contents
Acknowledgements ................................................................................................ 1
Abstract ............................................................................................................... 2
1. Introduction .............................................................................................. 3
2. History of NUTS, changes between classifications ........................................... 4
3. Conversion from NUTS 2010 to NUTS 2006 and vice versa .............................. 6
3.1 Conversion from NUTS 2010 to NUTS 2006 ............................................... 6
3.2 Conversion from NUTS 2006 to NUTS 2010 ............................................. 12
4. Conclusion .............................................................................................. 13
References ......................................................................................................... 14
Annex ................................................................................................................ 15
List of abbreviations and definitions ....................................................................... 20
List of figures ...................................................................................................... 21
List of tables ....................................................................................................... 22
1
Acknowledgements
The author acknowledges the comments and suggestions from Francesco Di Comite and
thanks all the members of the REMO team for their collaborative working.
2
Abstract
This technical report presents the methodology followed to transform regional data
between different NUTS classifications in RHOMOLO, the spatial computable general
equilibrium model developed by the European Commission to evaluate the impact of
Cohesion Policy. This method has been designed for the conversion between NUTS 2006
and NUTS 2010 in both directions, but the same philosophy can be extended to
transform data between any pair of NUTS classifications. It has been applied to the
construction of two regional databases for RHOMOLO-v2 in 2010, one covering the EU-
27 regions according to NUTS 2006, the other for the EU-28 regions according to NUTS
2010.
Keywords: NUTS, EU-28, regional database.
JEL Codes: D57, E16, R10.
3
1. Introduction
RHOMOLO-v2 is a spatial computable general equilibrium model developed by the
Directorate General Joint Research Centre (DG JRC) and the Directorate General for
Regional and Urban Policy (DG REGIO) of the European Commission. This model has
been designed to evaluate the effects of Cohesion Policy on economic growth for all the
regions of the European Union. In fact, RHOMOLO has already been used to assess the
impact of Cohesion Policy for the 2007-2013 Programming period and will be used for
the ex-ante impact assessment for the period 2014-2020. A full description of the model
can be found in Mercenier et al. (2016).
In order to capture the characteristics of regions and interregional connections, the
database has been built at the NUTS 2 level. The Nomenclature of Territorial Units for
Statistics, abbreviated NUTS (from the French version Nomenclature des unités
territoriales statistiques), is a hierarchical classification system to divide the EU territory
for the purpose of collection, development and harmonisation of EU regional statistics,
and socio-economic analyses of the regions and for the framing of EU regional policies.
There are three levels -NUTS 1, 2 and 3 respectively, moving from larger to smaller
territorial units-. NUTS 2 levels correspond to regions for the application of regional
policies, Eurostat (2011). The first NUTS was adopted in May 2003 –NUTS 2003- and
since then three major revisions have been implemented –NUTS 2006, NUTS 2010 and
NUTS 2013-.
A number of data sources are used by RHOMOLO to apprehend the regional and sectoral
characteristics and spatial connections in the base year 2010: a set of 28 national Social
accounting matrices (SAMs) has been built for the EU-28, and further regionalised at
NUTS 2 level, to capture the circular flow of income for the economy in the regions; an
interregional transport cost matrix is used to estimate iceberg-type transport cost
between regions; the interregional trade flows show the spatial connections; a set of
elasticities model the choices made by the economic agents; the Herfindahl indices are
used to estimate the number of firms at country-sector level.
The main sources for the construction of the SAMs at national level are the Supply and
use tables from the World Input Output Database and National Accounts from Eurostat
(Álvarez-Martínez and López-Cobo, 2016). For the regionalisation process a non-survey
method is applied, making extensive use of regional data available from Eurostat, as well
as interregional trade flows data. A description of the methodology can be found in
López-Cobo (2016). The data used to build the regional SAMs come from several
sources, often developed under different sectoral (activity) or regional classifications.
The available regional data from Eurostat for the year 2010 follows the NUTS 2010
classification; while the interregional trade flows have been built under the NUTS 2006
version, based on Thissen et al. (2015). Ultimately, the Cohesion Funds allocated during
the 2007-2013 Programming period referred to NUTS 2006 regions, which explains the
initial development of the database for RHOMOLO under the NUTS 2006 version. On the
other hand, the funds programmed for the period 2014-2020 will be allocated to the
NUTS 2010 regions, so both classifications must be used.
Therefore, a conversion procedure between different versions of the regional
classification was needed. This report describes the procedure followed to convert data
from NUTS 2010 to NUTS 2006 and backward, so that RHOMOLO-v2 can be used to
assess the impact of the Cohesion policy irrespective of the programming period or
regional classification. This is not a research paper but rather a technical note, intended
for empirical economists dealing with data from different NUTS classifications. This
paper’s primary objective is to describe the decisions taken in the data conversion, as
well as to stimulate and promote discussion rather than to provide a definitive solution.
4
2. History of NUTS, changes between classifications
The Commission Regulation (EC) 1059/2003 gave for the first time NUTS a legal status,
after around thirty years of implementation and updating of the NUTS classification
under a series of "gentlemen's agreements" between the Member States and Eurostat.
The regulation also requires the stability of the classification for at least three years.
Stability makes sure that data refers to the same regional unit for a certain period of
time. This is crucial for statistics, in particular for time-series. However, sometimes
national interests require changing the regional breakdown of a country. When this
happens the country concerned informs the European Commission about the changes.
The Commission in turn amends the classification at the end of period of stability
according to the rules of the NUTS Regulation.
Figure 1. Evolution of NUTS classifications
A first regular amendment to the annexes was adopted by Commission Regulation (EC)
No 105/2007 and the NUTS version 2003 was replaced by version 2006 on 1 January
20081. The second regular amendment to the annexes was adopted by Commission
Regulation (EU) No 31/2011, NUTS 2010. The third regular amendment to the annexes
was adopted by Commission Regulation (EU) No 1319/2013, NUTS 20132.
According to the regulations, in case of an amendment to the classification, the Member
State concerned has to replace historical data by time series according to the new
regional breakdown within two years. In such cases, the time series is substituted by
one updated according to the newest classification, and the data following the previous
classification are not available anymore from the Eurostat’s website, not even under
request. Due to this break in the series, if there is a need to use data according to and
old version of the NUTS, a conversion procedure has to be applied. This conversion is
sometimes straightforward, but in some cases more detailed information needs to be
used.
There are mainly three types of changes that affect data: regions can merge, split or
change boundaries. Changes in region names or codes do not affect data integrity and
therefore are disregarded in this exercise. Table 1 summarises the changes between
NUTS 2006 and NUTS 2010 that need to be addressed. Table A 1 in Annex provides the
1 This was preceded by completions of the NUTS classification with the regional breakdowns of the countries that joined the EU in 2004 and 2007 (see Commission Regulation (EC) No 1888/2005 and Commission Regulation (EC) No 176/2008). 2 The fourth, extraordinary amendment to the annexes was adopted by Commission Regulation (EU) No 868/2014, entered into force on 8 August 2014 and applicable, with regard to the
transmission of data to the Commission (Eurostat), from 1 January 2016.
Source: Eurostat
5
list of the 267 NUTS 2006 regions covered by RHOMOLO. Table A 2 in Annex shows the
complete list of changes at NUTS 2 level from NUTS 2006 to NUTS 2010.
Table 1. Regions with changes between 2006 and 2010
Change from 2006 to 2010
NUTS 2006 code (old)
NUTS 2010 code (new)
Explanation (new = old)
1. Split FI18 FI1B, FI1C FI1B + FI1C = FI18, recalculation by NSI
2. Merge DE41, DE42 DE40 DE40 = DE41 + DE42
FI13, FI1A FI1D FI1D = FI13 + FI1A
3. Boundary shift
DED1, DED3 DED4, DED5
recalculation by NSI ITD5, ITE3 ITH5, ITI3
UKD2, UKD5 UKD6, UKD7
Source: Own elaboration based on Eurostat.
Figure 2 illustrates, on the left side, change type 1 –split- and type 2 -merge- with
Finnish regions, and on the right side, type –boundary shift- of British regions.
Figure 2. Map showing the changes between NUTS 2006 and NUTS 2010 in NUTS 2 regions of Finland and United Kingdom
Source: Eurostat: http://ec.europa.eu/eurostat/web/nuts/history
6
3. Conversion from NUTS 2010 to NUTS 2006 and vice versa
3.1 Conversion from NUTS 2010 to NUTS 2006
A specific procedure is put in place to address each of the three types of changes
described above. The procedure applied in each case differs, as does the accuracy of
data obtained.
When a region splits into two regions, like is the case of FI18, Etelä-Suomi, which splits
into FI1B, Helsinki-Uusimaa, and FI1C, Etelä-Suomi, the simple aggregation of data of
these two NUTS 2010 regions provides the numbers corresponding to the original NUTS
2006 region. Whatever regional data we deal with, by aggregating we can reconstruct
the original region data, not an estimation.
In the case of merged regions (for example the new FI1D, Pohjois- ja Itä-Suomi in
2010, is the aggregation of the old FI13, Itä-Suomi, and FI1A, Pohjois-Suomi in 2006),
we need to use more detailed information to estimate data for the two old regions. By
comparing the 2006 and 2010 versions of the classification, we can see that at the
NUTS-3 level there is full correspondence between the two old regions and the new
merged region (Table 2). In our example, we know that the four new NUTS 3 regions
FI1D1 to FI1D4 match with the four old NUTS 3 regions composing FI13 (FI131 to
FI134) and that the new FI1D5 to FI1D7 match with the old FI1A1 to FI1A3, with only
code changes.
Table 2. Correspondence at NUTS 3 level in merged regions from 2006 to 2010. The case
of Finnish regions
Code 2006
Code 2010 NUTS level 2 NUTS level 3 Change
FI13 FI1D (part) Itä-Suomi Merged
FI131 FI1D1 Etelä-Savo Code change
FI132 FI1D2 Pohjois-Savo Code change
FI133 FI1D3 Pohjois-Karjala Code change
FI134 FI1D4 Kainuu Code change
FI1A FI1D (part) Pohjois-Suomi Merged
FI1A1 FI1D5 Keski-Pohjanmaa Code change
FI1A2 FI1D6 Pohjois-Pohjanmaa Code change
FI1A3 FI1D7 Lappi Code change
Source: Eurostat.
Note: Labels correspond to version 2010, except if in italics (version 2006).
Note: NUTS 2010 codes in italics and with the note "(part)" are created from whole NUTS 2006 regions. Data can thus be aggregated from NUTS 2006 to NUTS 2010.
Hence, we are able to compute data for the NUTS 2 regions according to NUTS 2006
subject that we have it at NUTS 3 level according to NUTS 2010. We collect NUTS 2010
GVA data at NUTS 3 level broken down by sector3 and reconstruct NUTS 2006 GVA data
at NUTS 2 level for the same year by aggregation. To illustrate the procedure, we use
total GVA not disaggregated by sector (Table 3). By doing this we have computed the
exact GVA figure for the old regions FI13 and FI1A. However our aim is to obtain other
regional data for these regions but, unfortunately, the availability of NUTS 3 level data in
Eurostat is very scarce. As a consequence, we need to estimate data for the old regions
3 Eurostat (2014). Regional economic accounts (ESA95). Gross value added at basic prices by
NUTS 3 regions (NACE Rev. 2) (nama_r_e3vab95r2) [Data file]. Downloaded on 2014 July 29 from
http://ec.europa.eu/eurostat/data/database. Not available anymore.
7
using their GVA data as a proxy. To this end, we compute the GVA shares of the NUTS-2
regions existing under version 2006 (the share of each individual region over the merged
one), and multiply these shares by the regional data of the new aggregated region FI1D.
In our example, the GVA share of FI13 over FI1D is 47.49%; hence, we estimate the
regional indicators other than value added needed for the SAM of FI13, such as
compensation of employees, gross fixed capital formation, household accounts, etc., by
multiplying these indicators observed for FI1D by this share. Similarly, we get data for
FI1A. Here, the source of error primarily lies on the different relation that may exist in
the economic variable between two regions, say compensation of employees, with
respect to their GVA. If, for example, region FI13 had a higher productivity level than
FI1A (computed as GVA over employment), and assuming unity salary over the regions
(employment equals compensation of employees), we would be overestimating the
compensation of employees in FI13 by applying the GVA share. In this case, other
regional proxies showing the different productivity levels might be used to improve this
estimation, but again we face the problem of data availability when it comes to NUTS 3
or even NUTS 2 level data.
Table 3. Reconstruction of GVA at NUTS 2 level according to NUTS 2006. Merged regions.
Year 2010
NUTS 2010 NUTS 2006
NUTS 2 GVA 2010 NUTS 3 GVA 2010 NUTS 3 NUTS 2 GVA 2010 GVA share
DE40 49,002
DE403 1,768 DE411
DE41 19,813 0.404323
DE405 2,583 DE412
DE409 2,796 DE413
DE40A 3,669 DE414
DE40C 3,119 DE415
DE40D 1,806 DE416
DE40F 1,443 DE417
DE40I 2,630 DE418
DE401 1,619 DE421
DE42 29,190 0.595677
DE402 2,482 DE422
DE404 4,765 DE423
DE406 4,267 DE424
DE407 1,798 DE425
DE408 2,075 DE426
DE40B 2,049 DE427
DE40E 3,215 DE428
DE40G 3,607 DE429
DE40H 3,313 DE42A
FI1D 30,439
FI1D1 3,359 FI131
FI13 14,456 0.474905 FI1D2 5,872 FI132
FI1D3 3,527 FI133
FI1D4 1,698 FI134
FI1D5 1,795 FI1A1
FI1A 15,983 0.525095 FI1D6 9,769 FI1A2
FI1D7 4,419 FI1A3
Source: Own elaboration.
8
Following this procedure at the sector level we obtain the GVA of the old regions sector
(Table 4. Estimation of GVA by sector (NACE Rev.2) at NUTS 2 level according to NUTS
2006. Merged regions). These are not the final figures for GVA in these regions, since
there is still the need to transform data from NACE Rev. 2 (Eurostat data) to NACE Rev.
1 (WIOD data), as explained in López-Cobo (2016).
Table 4. Estimation of GVA by sector (NACE Rev.2) at NUTS 2 level according to NUTS
2006. Merged regions. Year 2010
NUTS 2006
NUTS 2 Total GVA A B-E F G-J K-N O-U
DE41 19,812.7 499.6 4,061.1 1,403.1 3,362.4 4,586.6 5,899.9
DE42 29,189.5 418.9 6,049.4 1,732.5 5,337.1 7,431.0 8,220.4
FI13 14,455.6 1,126.5 2,708.9 1,121.2 2,328.1 2,919.1 4,251.6
FI1A 15,983.3 665.4 3,823.5 1,211.7 2,739.5 3,109.4 4,431.7
GVA shares
DE41 0.4043 0.5439 0.4017 0.4475 0.3865 0.3817 0.4178
DE42 0.5957 0.4561 0.5983 0.5525 0.6135 0.6183 0.5822
FI13 0.4749 0.6287 0.4147 0.4806 0.4594 0.4842 0.4896
FI1A 0.5251 0.3713 0.5853 0.5194 0.5406 0.5158 0.5104
Source: Own elaboration.
In the case of boundary shifts, there are two different cases:
a) those where the reproducibility of NUTS 2006 regions from NUTS 2010 cannot be
reached despite the availability of NUTS 3 level data. This happens when at least one
NUTS 3 region changed its boundaries between both classifications, capturing now
part of the territory of another NUTS 3 region belonging to a different old NUTS 2
region. Therefore, since there is not full correspondence between NUTS 3 level
regions of 2006 and 2010 versions of the classification, the old NUTS 2 region cannot
be fully reproduced.
b) those where the reproducibility relies on the availability of information at NUTS 3
level, depending also on the direction in which we want to make the transformation,
that is, from NUTS 2006 to NUTS 2010 or backwards. This is the case where an entire
NUTS 3 region switches between two NUTS 2 regions but keeps its boundaries
unchanged, only the boundaries of the NUTS 2 regions change.
Case a) corresponds to the British and Italian regions that shifted boundaries from 2006
to 2010 (Table 1). Here we illustrate the case with the British regions. In Figure 2 (right)
and Table 5 we can see how the new region UKD7 (Merseyside) includes old UKD5
(Merseyside) plus part of UKD2 (Cheshire). The figure shows the NUTS 2010 regions in
black font and coloured areas, and the NUTS 2006 boundaries with red lines. The NUTS
3 regions UKD51 (East Merseyside) and UKD21 (Halton and Warrington) have shifted
boundaries by increasing their size (the former) to become UKD71 or reducing it (the
latter) to become UKD61. The green area inside UKD6 shows the part of UKD71 that had
belonged to UKD21 before. Data for the new NUTS 3 and NUTS 2 regions have been
recomputed by the National Statistics Institute, but no further information is available to
allow us to reconstruct the old regions. Therefore, on top of the source of error explained
in the case of the merged regions, here we face an extra issue due to the fact that old
NUTS 2 regions cannot be fully reconstructed even when NUTS 3 level data are
available. Here we would put in place a two-step procedure: we would first aggregate
data corresponding to the two new regions (UKD6 and UKD7) and disaggregate them
into the two old regions (UKD2 and UKD5) using the GVA shares. But since it is not
9
possible to have GVA data corresponding to the boundary shift, but only to the whole
new NUTS 3 region, we have to assume that the GVA of the NUTS 3 regions before
(UKD21) and after the boundary shift (UKD61) are the same. Therefore, by neglecting
the boundary shift of the NUTS 3 region, what we are assuming in practical terms is that
the NUTS 2 regions have not changed. Consequently, NUTS 2 regions where one or
several NUTS 3 regions have seen a boundary shift are not recomputed in fact, they stay
untouched and data for the new NUTS 2010 region UKD7 are assumed to be the same as
for the old NUTS 2006 region UKD5. Nevertheless, this boundary shift affects only one
NUTS3 region in each NUT2 region. Therefore we can neglect the boundary shift and
assume that data of the older region are approximately equal to data of the newer
region.
Table 5. Correspondence at NUTS 3 level in the case of regions that shift boundaries from 2006 to 2010. Case a) British regions
Code 2006 Code 2010 NUTS level 2 NUTS level 3 Change
UKD2 Cheshire Boundary shift
UKD6 Cheshire New region
UKD21 Halton and Warrington Boundary shift
UKD61 Warrington New region
UKD22 Cheshire CC Split
UKD22 (part) UKD62 Cheshire East New region
UKD22 (part) UKD63 Cheshire West and Chester
New region
UKD5 Merseyside Boundary shift
UKD7 Merseyside
New region
UKD51 East Merseyside Boundary shift
UKD71 East Merseyside New region
UKD52 UKD72 Liverpool Code change
UKD53 UKD73 Sefton Code change
UKD54 UKD74 Wirral Code change
Source: Eurostat.
Note: Labels correspond to version 2010, except if in italics (version 2006).
Note: NUTS 2006 codes in italics and with the note "(part)" are split in a number of whole NUTS 2010 regions.
Similarly, the old ITD5 region (Emilia-Romagna) has a correspondence with the new
ITH5 region with the same name. The only difference between ITD5 and ITH5 lies in that
its ninth NUTS 3 region, ITD59 (Rimini), experienced a boundary shift, being
transformed into the bigger ITH59 (Rimini). This boundary shift is in turn compensated
by the complementary reduction in the old ITE31 (Pessaro e Urbino), which was part of
the old NUTS 2 ITE3 (Marche), which are converted into the new ITI31 (Pessaro e
Urbino) and ITI3 (Marche) respectively.
Case b) is represented by the German regions DED1 (Chemnitz) and DED3 (Leipzig)
being transformed into the new DED4 and DED5 with same names. As shown by Table 6,
at NUTS 3 level there are a few old regions merging into new ones, but at NUTS 2 level
the only significant difference lies in the NUTS 3 region DED33 leaving the old NUTS 2
region DED3 to become part of the new DED4 (old DED1). Since DED33 does not
become an entire new NUTS 3 region, but only a part of the new DED43, there is not full
correspondence at NUTS 3 level between classifications. Consequently, even if we had
NUTS 3 level data for the new regions, we would not be able to reconstruct the old NUTS
2 regions. However, if we were interested in the conversion in the other direction, from
NUTS 2006 to NUTS 2010, we would be able to reconstruct entirely data for the NUTS 2
new regions using NUTS 3 level data of NUTS 2006 regions.
10
Table 6. Correspondence at NUTS 3 level in the case of regions that shift boundaries from 2006 to 2010. Case b) German regions
Code 2006 Code 2010 NUTS level 2 NUTS level 3 Change
DED1 Chemnitz Boundary shift
DED4 Chemnitz New region
DED11 DED41 Chemnitz, Kreisfreie Stadt Code change
DED12 DED44 (part) Plauen, Kreisfreie Stadt Merged
DED13 DED45 (part) Zwickau, Kreisfreie Stadt Merged
DED14 DED42 (part) Annaberg Merged
DED15 DED45 (part) Chemnitzer Land Merged
DED16 DED43 (part) Freiberg Merged
DED17 DED44 (part) Vogtlandkreis Merged
DED18 DED42 (part) Mittlerer Erzgebirgskreis Merged
DED19 DED43 (part) Mittweida Merged
DED1A DED42 (part) Stollberg Merged
DED1B DED42 (part) Aue-Schwarzenberg Merged
DED1C DED45 (part) Zwickauer Land Merged
DED42 Erzgebirgskreis New region
DED43 Mittelsachsen New region
DED44 Vogtlandkreis New region
DED45 Zwickau New region
DED3 Leipzig Boundary shift
DED5 Leipzig New region
DED31 DED51 Leipzig, Kreisfreie Stadt Code change
DED32 DED53 (part) Delitzsch Merged
DED33 DED43 (part) Döbeln Merged
DED34 DED52 (part) Leipziger Land Merged
DED35 DED52 (part) Muldentalkreis Merged
DED36 DED53 (part) Torgau-Oschatz Merged
DED52 Leipzig New region
DED53 Nordsachsen New region
Source: Eurostat.
Note: NUTS 2010 codes in italics and with the note "(part)" are created from whole NUTS 2006 regions. Data can thus be aggregated from NUTS 2006 to NUTS 2010.
To estimate data for the old regions, DED1 and DED3, we can choose whether to assume
that the contribution of the NUTS 3 region DED33 that switches between NUTS 2 regions
is negligible, in which case we assume that data for the old DED1 region are the same as
those for the new DED4 region, and DED3 equals DED5; or we can estimate the amount
of DED4 figures due to DED33 and add it to DED5 data to approximate the old region
DED3. Table 7 shows the composition of the 2010 NUTS 2 and NUTS 3 regions in terms
11
of 2006 NUTS 3 regions and the estimation of their GVA. Due to the lack of data at NUTS
3 level according to the 2006 classification, we can only assume that the three NUTS 3
regions constituting DED43 contribute equally to the GVA of the latter. This way, the
estimated GVA of DED33 is one third of DED43 GVA. Now we can deduct the figure for
DED33 (2,201.5) from DED4 (29,377) to estimate DED1 GVA (27,175), and similarly,
add it to DED5 GVA to get an approximation to DED3 GVA. For indicators different than
GVA, we can use the GVA ratios (old region’s GVA over new region’s GVA) to obtain a
multiplier (0.925060) applicable to other economic indicators of the NUTS 2010 regions
to get an estimation of their NUTS 2006 counter parts. As explained before, we are
relying on the assumption that all economic variables are proportional to GVA or, in
other words, the relationship between one economic variable in two regions is
proportional to their GVA.
Table 7. Estimation of GVA at NUTS 2 level according to NUTS 2006. Regions with boundaries shift. Case b) German regions. Year 2010
NUTS 2010 NUTS 2006
NUTS 2 GVA 2010
NUTS 3 GVA 2010
NUTS 3 Estimated GVA 2010
NUTS 2 Estimated GVA 2010
GVA ratio (NUTS 2006/ NUTS
2010)
DED4 29,377
DED41 5,904 DED11 5,903.8
DED1= DED4 - DED33
27,175
0.925060
DED42 5,837
DED14 1,459.1
DED18 1,459.1
DED1A 1,459.1
DED1B 1,459.1
DED43 6,604
DED16 2,201.5
DED19 2,201.5
DED33 2,201.5
DED44 4,251 DED12 2,125.6
DED17 2,125.6
DED45 6,781
DED13 2,260.3
DED15 2,260.3
DED1C 2,260.3
DED5 20,887
DED51 12,357 DED31 12,357.0
DED3 = DED5 + DED33
23,088
1.105399
DED52 4,690 DED34 2,345.2
DED35 2,345.2
DED53 3,840 DED32 1,919.8
DED36 1,919.8
Source: Own elaboration.
Following this procedure at the sector level we estimate the GVA of the old regions DED1
and DED3 by sector (Table 8). These are not the final figures for GVA in these regions,
since there is still the need to transform data from NACE Rev. 2 (Eurostat data) to NACE
Rev. 1 (WIOD data), as explained in López-Cobo, 2016.
12
Table 8. Estimation of GVA by sector (NACE Rev.2) at NUTS 2 level according to NUTS 2006. Regions with boundary shift. Case b) German regions. Year 2010
NUTS 2006
NUTS 2 and NUTS 3
Total GVA A B-E F G-J K-N O-U
DED1 27,175.1 253.6 7,273.9 1,954.2 4,630.8 5,861.9 7,200.9
DED11 5,903.8 6.0 1,087.2 364.4 1,079.8 1,517.0 1,849.3
DED14 1,459.1 17.1 400.6 123.6 236.6 290.9 390.3
DED18 1,459.1 17.1 400.6 123.6 236.6 290.9 390.3
DED1A 1,459.1 17.1 400.6 123.6 236.6 290.9 390.3
DED1B 1,459.1 17.1 400.6 123.6 236.6 290.9 390.3
DED16 2,201.5 39.3 684.4 154.9 382.9 447.8 492.2
DED19 2,201.5 39.3 684.4 154.9 382.9 447.8 492.2
DED12 2,125.6 23.7 581.0 166.3 348.4 431.7 574.6
DED17 2,125.6 23.7 581.0 166.3 348.4 431.7 574.6
DED13 2,260.3 17.7 684.5 150.9 380.7 474.2 552.3
DED15 2,260.3 17.7 684.5 150.9 380.7 474.2 552.3
DED1C 2,260.3 17.7 684.5 150.9 380.7 474.2 552.3
DED3 23,088.4 259.9 4,679.5 1,567.5 4,746.7 5,413.4 6,421.3
DED31 12,357.0 7.5 1,687.2 657.7 2,656.0 3,404.7 3,943.9
DED34 2,345.2 40.2 753.5 215.6 380.2 428.9 527.0
DED35 2,345.2 40.2 753.5 215.6 380.2 428.9 527.0
DED32 1,919.8 66.4 400.5 161.9 473.7 351.6 465.7
DED36 1,919.8 66.4 400.5 161.9 473.7 351.6 465.7
DED33 2,201.5 39.3 684.4 154.9 382.9 447.8 492.2
Ratio New region / Old region
DED1 / DED4 0.9251 0.8659 0.9140 0.9266 0.9236 0.9290 0.9360
DED3 / DED5 1.1054 1.1775 1.1713 1.1097 1.0877 1.0902 1.0830
Source: Own elaboration.
3.2 Conversion from NUTS 2006 to NUTS 2010
In order to use RHOMOLO for the ex-ante impact assessment of Cohesion Policy 2014-
2020, the database needs to be built under (or converted to) the 2010 version of the
NUTS classification. In this case, we can use the regional data from Eurostat as it is but
we need to convert the inter-regional trade flows and the inter-regional transport cost
matrix.
For the trade flows it makes sense to apply the same underlying strategy explained in
the previous section. We rely on the GVA shares of regions to disaggregate trade data
for regions that split or shifted boundaries between 2006 and 2010. When a region
splits into two regions, like is the case of FI18, Etelä-Suomi (Table 1), splitting into FI1B
and FI1C, we break down the trade data according to the GVA shares of the new
regions.
For merged regions, like FI1D becoming the union of FI13 and FI1A, the simple
aggregation of trade data of these two NUTS 2006 regions provides the trade
corresponding to the new NUTS 2010 region.
13
4. Conclusion
This report describes the methodology designed to convert regional data between
different NUTS classifications, as applied in RHOMOLO. It fills a gap in the definition of
an applied approach to use regional data, when there is a need for data from an
outdated NUTS classification. The importance to define a methodology lies in the fact
that statistics following the previous classifications are not available from the Eurostat’s
website. The availability of data at NUTS 2 and NUTS 3 level is a key factor for the
development of the method. Consequently, the results are more robust for GVA than for
other economic variables such as employment or gross fix capital formation, for which
NUTS 3 level data are not available. Possible improvements of the proposed
methodology proposed include the identification of proxies to adjust the relationship
between GVA and other variables at the maximum level of disaggregation.
14
References
Álvarez-Martínez M.T. and López-Cobo M. (2016), "Social Accounting Matrices for the
EU-27 in 2010. Building a new database for RHOMOLO". Institute for Prospective
Technological Studies, DG-JRC, European Commission; JRC101673.
Mercenier, J., Álvarez-Martinez, M., Brandsma, A., Di Comite, F., Diukanova, O., Kancs,
d’A., Lecca, P., López-Cobo, M., Monfort, Ph., Persyn, D., Rillaers, A., Thissen M. and
Torfs, W. (2016). "RHOMOLO v2 Model Description: A spatial computable general
equilibrium model for EU regions and sectors," JRC Technical reports JRC100011,
European Commission, DG Joint Research Centre, EUR 27728 EN, doi:10.2791/18446.
Eurostat (2011), “Regions in the European Union. Nomenclature of territorial units for
statistics. NUTS 2010/EU-27”, Luxembourg.
López-Cobo (2016), "Regionalisation of Social Accounting Matrices for the EU-28 in
2010. A regional database for RHOMOLO at NUTS 2 level", European Commission, DG
Joint Research Centre, JRC.B.3, Seville; JRC104029.
15
Annex
Table A 1 NUTS-2 Regions in RHOMOLO according to NUTS 2006 classification
Code Name
BE10 Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest
BE21 Prov. Antwerpen
BE22 Prov. Limburg (BE)
BE23 Prov. Oost-Vlaanderen
BE24 Prov. Vlaams-Brabant
BE25 Prov. West-Vlaanderen
BE31 Prov. Brabant Wallon
BE32 Prov. Hainaut
BE33 Prov. Liège
BE34 Prov. Luxembourg (BE)
BE35 Prov. Namur
BG31 Северозападен
BG32 Северен централен
BG33 Североизточен
BG34 Югоизточен
BG41 Югозападен
BG42 Южен централен
CZ01 Praha
CZ02 Střední Čechy
CZ03 Jihozápad
CZ04 Severozápad
CZ05 Severovýchod
CZ06 Jihovýchod
CZ07 Střední Morava
CZ08 Moravskoslezsko
DK01 Hovedstaden
DK02 Sjælland
DK03 Syddanmark
DK04 Midtjylland
DK05 Nordjylland
DE11 Stuttgart
DE12 Karlsruhe
DE13 Freiburg
DE14 Tübingen
DE21 Oberbayern
DE22 Niederbayern
DE23 Oberpfalz
DE24 Oberfranken
DE25 Mittelfranken
DE26 Unterfranken
Code Name
DE27 Schwaben
DE30 Berlin
DE41 Brandenburg - Nordost
DE42 Brandenburg - Südwest
DE50 Bremen
DE60 Hamburg
DE71 Darmstadt
DE72 Gießen
DE73 Kassel
DE80 Mecklenburg-Vorpommern
DE91 Braunschweig
DE92 Hannover
DE93 Lüneburg
DE94 Weser-Ems
DEA1 Düsseldorf
DEA2 Köln
DEA3 Münster
DEA4 Detmold
DEA5 Arnsberg
DEB1 Koblenz
DEB2 Trier
DEB3 Rheinhessen-Pfalz
DEC0 Saarland
DED1 Chemnitz
DED2 Dresden
DED3 Leipzig
DEE0 Sachsen-Anhalt
DEF0 Schleswig-Holstein
DEG0 Thüringen
EE00 Eesti
IE01 Border, Midland and Western
IE02 Southern and Eastern
GR11 Aνατολική Μακεδονία, Θράκη
GR12 Κεντρική Μακεδονία
GR13 Δυτική Μακεδονία
GR14 Θεσσαλία
GR21 Ήπειρος
GR22 Ιόνια Νησιά
GR23 Δυτική Ελλάδα
GR24 Στερεά Ελλάδα
16
Code Name
GR25 Πελοπόννησος
GR30 Aττική
GR41 Βόρειο Αιγαίο
GR42 Νότιο Αιγαίο
GR43 Κρήτη
ES11 Galicia
ES12 Principado de Asturias
ES13 Cantabria
ES21 País Vasco
ES22 Comunidad Foral de Navarra
ES23 La Rioja
ES24 Aragón
ES30 Comunidad de Madrid
ES41 Castilla y León
ES42 Castilla-La Mancha
ES43 Extremadura
ES51 Cataluña
ES52 Comunidad Valenciana
ES53 Illes Balears
ES61 Andalucía
ES62 Región de Murcia
ES63 Ciudad Autónoma de Ceuta
ES64 Ciudad Autónoma de Melilla
ES70 Canarias
FR10 Île de France
FR21 Champagne-Ardenne
FR22 Picardie
FR23 Haute-Normandie
FR24 Centre
FR25 Basse-Normandie
FR26 Bourgogne
FR30 Nord - Pas-de-Calais
FR41 Lorraine
FR42 Alsace
FR43 Franche-Comté
FR51 Pays de la Loire
FR52 Bretagne
FR53 Poitou-Charentes
FR61 Aquitaine
FR62 Midi-Pyrénées
FR63 Limousin
FR71 Rhône-Alpes
FR72 Auvergne
FR81 Languedoc-Roussillon
Code Name
FR82 Provence-Alpes-Côte d'Azur
FR83 Corse
ITC1 Piemonte
ITC2 Valle d'Aosta/Vallée d'Aoste
ITC3 Liguria
ITC4 Lombardia
ITD1 Provincia Autonoma di Bolzano/Bozen
ITD2 Provincia Autonoma di Trento
ITD3 Veneto
ITD4 Friuli-Venezia Giulia
ITD5 Emilia-Romagna
ITE1 Toscana
ITE2 Umbria
ITE3 Marche
ITE4 Lazio
ITF1 Abruzzo
ITF2 Molise
ITF3 Campania
ITF4 Puglia
ITF5 Basilicata
ITF6 Calabria
ITG1 Sicilia
ITG2 Sardegna
CY00 Κύπρος
LV00 Latvija
LT00 Lietuva
LU00 Luxembourg
HU10 Közép-Magyarország
HU21 Közép-Dunántúl
HU22 Nyugat-Dunántúl
HU23 Dél-Dunántúl
HU31 Észak-Magyarország
HU32 Észak-Alföld
HU33 Dél-Alföld
MT00 Malta
NL11 Groningen
NL12 Friesland (NL)
NL13 Drenthe
NL21 Overijssel
NL22 Gelderland
NL23 Flevoland
NL31 Utrecht
NL32 Noord-Holland
17
Code Name
NL33 Zuid-Holland
NL34 Zeeland
NL41 Noord-Brabant
NL42 Limburg (NL)
AT11 Burgenland (AT)
AT12 Niederösterreich
AT13 Wien
AT21 Kärnten
AT22 Steiermark
AT31 Oberösterreich
AT32 Salzburg
AT33 Tirol
AT34 Vorarlberg
PL11 Łódzkie
PL12 Mazowieckie
PL21 Małopolskie
PL22 Śląskie
PL31 Lubelskie
PL32 Podkarpackie
PL33 Świętokrzyskie
PL34 Podlaskie
PL41 Wielkopolskie
PL42 Zachodniopomorskie
PL43 Lubuskie
PL51 Dolnośląskie
PL52 Opolskie
PL61 Kujawsko-Pomorskie
PL62 Warmińsko-Mazurskie
PL63 Pomorskie
PT11 Norte
PT15 Algarve
PT16 Centro (PT)
PT17 Lisboa
PT18 Alentejo
PT20 Região Autónoma dos Açores
PT30 Região Autónoma da Madeira
RO11 Nord-Vest
RO12 Centru
RO21 Nord-Est
RO22 Sud-Est
RO31 Sud - Muntenia
RO32 Bucureşti - Ilfov
RO41 Sud-Vest Oltenia
Code Name
RO42 Vest
SI01 Vzhodna Slovenija
SI02 Zahodna Slovenija
SK01 Bratislavský kraj
SK02 Západné Slovensko
SK03 Stredné Slovensko
SK04 Východné Slovensko
FI13 Itä-Suomi
FI18 Etelä-Suomi
FI19 Länsi-Suomi
FI1A Pohjois-Suomi
FI20 Åland
SE11 Stockholm
SE12 Östra Mellansverige
SE21 Småland med öarna
SE22 Sydsverige
SE23 Västsverige
SE31 Norra Mellansverige
SE32 Mellersta Norrland
SE33 Övre Norrland
UKC1 Tees Valley and Durham
UKC2 Northumberland and Tyne and Wear
UKD1 Cumbria
UKD2 Cheshire
UKD3 Greater Manchester
UKD4 Lancashire
UKD5 Merseyside
UKE1 East Yorkshire and Northern Lincolnshire
UKE2 North Yorkshire
UKE3 South Yorkshire
UKE4 West Yorkshire
UKF1 Derbyshire and Nottinghamshire
UKF2 Leicestershire, Rutland and Northamptonshire
UKF3 Lincolnshire
UKG1 Herefordshire, Worcestershire and Warwickshire
UKG2 Shropshire and Staffordshire
UKG3 West Midlands
UKH1 East Anglia
UKH2 Bedfordshire and Hertfordshire
UKH3 Essex
18
Code Name
UKI1 Inner London
UKI2 Outer London
UKJ1 Berkshire, Buckinghamshire and Oxfordshire
UKJ2 Surrey, East and West Sussex
UKJ3 Hampshire and Isle of
Wight
UKJ4 Kent
UKK1 Gloucestershire, Wiltshire
and Bristol/Bath area
UKK2 Dorset and Somerset
UKK3 Cornwall and Isles of Scilly
Code Name
UKK4 Devon
UKL1 West Wales and The Valleys
UKL2 East Wales
UKM2 Eastern Scotland
UKM3 South Western Scotland
UKM5 North Eastern Scotland
UKM6 Highlands and Islands
UKN0 Northern Ireland
Source: Mercenier et al., 2016
Table A 2 Changes from NUTS 2006 to NUTS 2010 at NUTS-2 level
Code
2006 Code 2010 Label Change
Explanation (new =
old)
DE41 DE40 (part) Brandenburg - Nordost
Merged
DE42 DE40 (part) Brandenburg - Südwest
Merged
DE40 Brandenburg New region DE40 = DE41 + DE42
DED1
Chemnitz Boundary shift
DED4 Chemnitz New region recalculation by NSI
DED3
Leipzig Boundary shift
DED5 Leipzig New region recalculation by NSI
GR11 EL11 Aνατολική Μακεδονία, Θράκη
Code change EL11 = GR11
GR12 EL12 Κεντρική Μακεδονία Code change EL12 = GR12
GR13 EL13 Δυτική Μακεδονία Code change EL13 = GR13
GR14 EL14 Θεσσαλία Code change EL14 = GR14
GR21 EL21 Ήπειρος Code change EL21 = GR21
GR22 EL22 Ιόνια Νησιά Code change EL22 = GR22
GR23 EL23 Δυτική Ελλάδα Code change EL23 = GR23
GR24 EL24 Στερεά Ελλάδα Code change EL24 = GR24
GR25 EL25 Πελοπόννησος Code change EL25 = GR25
GR30 EL30 Aττική Code change EL30 = GR30
GR41 EL41 Βόρειο Αιγαίο Code change EL41 = GR41
GR42 EL42 Νότιο Αιγαίο Code change EL42 = GR42
GR43 EL43 Κρήτη Code change EL43 = GR43
GRZZ ELZZ Extra-Regio NUTS 2 Code change, label change
ELZZ = GRZZ
ITD1 ITH1 Provincia Autonoma di Bolzano/Bozen
Code change, label change
ITH1 = ITD1
ITD2 ITH2 Provincia Autonoma di Trento
Code change, label change
ITH2 = ITD2
ITD3 ITH3 Veneto Code change ITH3 = ITD3
19
Code
2006 Code 2010 Label Change
Explanation (new =
old)
ITD4 ITH4 Friuli-Venezia Giulia Code change ITH4 = ITD4
ITE4 ITI4 Lazio Code change ITI4 = ITE4
ITE1 ITI1 Toscana Code change ITI1 = ITE1
ITE2 ITI2 Umbria Code change ITI2 = ITE2
ITD5
Emilia-Romagna Boundary shift
ITH5 Emilia-Romagna New region recalculation by NSI
ITE3
Marche Boundary shift
ITI3 Marche New region recalculation by NSI
FI13 FI1D (part) Itä-Suomi Merged
FI1A FI1D (part) Pohjois-Suomi Merged
FI1D
Pohjois- ja Itä-Suomi
New region FI1D = FI13 + FI1A
FI18
Etelä-Suomi Split
FI18
(part) FI1B Helsinki-Uusimaa New region
FI1B + FI1C = FI18, recalculation by NSI
FI18
(part) FI1C Etelä-Suomi New region
FI1B + FI1C = FI18, recalculation by NSI
UKD2
Cheshire Boundary shift
UKD5
Merseyside Boundary shift
UKD6 Cheshire New region recalculation by NSI
UKD7 Merseyside New region recalculation by NSI
20
List of abbreviations and definitions
EU European Union
GVA Gross value added
NACE Statistical classification of economic activities in the European Community
NUTS Nomenclature of Territorial Units for Statistics
SAM Social accounting matrix
WIOD World input-output database
21
List of figures
Figure 1. Evolution of NUTS classifications ................................................................ 4
Figure 2. Map showing the changes between NUTS 2006 and NUTS 2010 in NUTS 2
regions of Finland and United Kingdom ...................................................... 5
22
List of tables
Table 1. Regions with changes between 2006 and 2010 ......................................... 5
Table 2. Correspondence at NUTS 3 level in merged regions from 2006 to 2010.
The case of Finnish regions ..................................................................... 6
Table 3. Reconstruction of GVA at NUTS 2 level according to NUTS 2006. Merged
regions. Year 2010 ................................................................................ 7
Table 4. Estimation of GVA by sector (NACE Rev.2) at NUTS 2 level according to
NUTS 2006. Merged regions. Year 2010 ................................................... 8
Table 5. Correspondence at NUTS 3 level in the case of regions that shift
boundaries from 2006 to 2010. Case a) British regions .............................. 9
Table 6. Correspondence at NUTS 3 level in the case of regions that shift
boundaries from 2006 to 2010. Case b) German regions .......................... 10
Table 7. Estimation of GVA at NUTS 2 level according to NUTS 2006. Regions
with boundaries shift. Case b) German regions. Year 2010 ....................... 11
Table 8. Estimation of GVA by sector (NACE Rev.2) at NUTS 2 level according to
NUTS 2006. Regions with boundary shift. Case b) German regions. Year
2010 .................................................................................................. 12
Annex
Table A 1 NUTS-2 Regions in RHOMOLO according to NUTS 2006 classification ........... 15
Table A 2 Changes from NUTS 2006 to NUTS 2010 at NUTS-2 level ........................... 18
Europe Direct is a service to help you find answers to your questions about the European Union
Free phone number (*): 00 800 6 7 8 9 10 11
(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.
A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server http://europa.eu
How to obtain EU publications
Our publications are available from EU Bookshop (http://bookshop.europa.eu),
where you can place an order with the sales agent of your choice.
The Publications Office has a worldwide network of sales agents.
You can obtain their contact details by sending a fax to (352) 29 29-42758.
XX-N
A-x
xxxx-E
N-N
doi:10.2791/039045
ISBN 978-92-79-64453-5
LF-N
A-2
8300-E
N-N